Conflicts on Prognosis Occur Over Half the Time Between Physicians and Surrogate Decision-Makers

EXECUTIVE SUMMARY

Conflicts between physicians and surrogate decision-makers involving the patient’s prognosis occur more than half the time, according to a recent study.

• Some conflicts involved differences in belief about the prognosis, as opposed to lack of information or misunderstandings.
• The need to maintain hope was a common reason for surrogates being more optimistic about prognosis than physicians.
• Misunderstandings stem from complex medical terminology used by physicians.

Conflicts between physicians and surrogate decision-makers involving the patient’s prognosis occur more than half the time, according to a recent study.¹

J. Randall Curtis, MD, MPH, one of the study’s authors and director of Cambia Palliative Care Center of Excellence at University of Washington School of Medicine in Seattle, says the extent of the discordance surprised him. “More than 50% of the time, there was a greater than 20% difference in prognostic estimates,” he notes.

Researchers conducted quantitative surveys and qualitative interviews at four ICUs at a major U.S. medical center, with 299 surrogate decision-makers and 99 physicians caring for 174 critically ill patients, from 2005 to 2009. Some key findings include the following:

• Physician-surrogate discordance occurred in 122 (53%) instances. Of these, 65 instances involved both misunderstandings by surrogates and differences in belief about the patient’s prognosis, 38 involved only misunderstandings, and seven were related to differences in belief only.
• Surrogates’ prognostic estimates were much more accurate than chance alone, but physicians’ prognostic estimates were statistically significantly more accurate than surrogates’.
• Among 71 surrogates interviewed who had beliefs about the prognosis that were more optimistic than that of the physician, the most common reasons were a need to maintain hope to benefit the patient, a belief that the patient had unique strengths unknown to the physician, and religious beliefs.

Difference in Opinion

Douglas B. White, MD, MAS, the study’s lead author, says, “The study’s findings point out very clearly that when doctors and families have different expectations about prognosis, it’s not always due to a misunderstanding.” White is the endowed chair for ethics in critical care medicine and director of the program on ethics and decision-making in critical illness at University of Pittsburgh Medical Center.

Sometimes, conflicts occurred simply because physicians and surrogates had different opinions. “It’s pretty clear that different judgments and beliefs were prevalent in this cohort,” says White.

The study’s findings “remind physicians and bioethicists that there’s a difference between the informed consent steps of understanding and appreciation,” says White. Surrogates can understand exactly what a doctor says, and still not appreciate how, and whether, it relates to the patient.

“For clinical ethics consultants, it may be worth conceptualizing part of their role as helping surrogates to think about how the prognosis does, or doesn’t, fit with their own perceptions,” says White. Bioethicists can explore why that difference exists. This could lead to further dialogue, allowing both parties to resolve the difference.

“Sometimes it will lead to the physician modifying his or her estimate, and sometimes it will lead to the family doing so,” says White.

The study highlights the importance of physicians exploring discrepancies in prognostic predictions and the reasons underlying those discrepancies, Curtis says. “In order for family members to be able to participate in surrogate decision-making for critically ill patients, it is important that we do our best to ensure they understand the patient’s prognosis,” he explains. This includes not just providing information, but also exploring beliefs that may cause discordance.

The study found that some conflicts were rooted in the family’s inability to face the reality of the prognosis.

White says, “That’s an area for ethicists to intervene on.” The ethicist can find ways to preserve hope that don’t rely on ignoring the gravity of the situation. “There are many different things that one can hope for, in the face of terminal illness, that don’t rely on the patient being magically cured,” says White.

Physicians often use complex language to talk about prognosis, and rarely check to see if surrogates understood what they said, found a 2009 study.² “They may focus on small improvements in kidney function, but don’t make it clear that things have not changed overall with regard to the patient’s prognosis,” says White, the study’s lead author.

Janice Firn, PhD, MSW, a clinical ethicist at the University of Michigan Health System, says there is a clear need for improvement in communication skills about prognosis, “from trainees all the way up.”

Clinicians often choose to use medical terminology, such as stating, “she has a poor prognosis,” instead of saying, “She will not survive. She is dying.”

“It’s easy to get used to the language we use in medicine, but it might not make sense to the layperson,” says Firn. “We often hear from the family, ‘If only I had known that.’”

In family meetings, clinicians often say, “He’s not doing very well.” Hearing this, says Firn, “The family might think, ‘Yes, that’s why we brought him to the hospital!’” They fail to realize that the clinician really means the patient is expected to die soon.

Clinicians may ask the surrogate, “What do you want us to do when she codes?” when they mean, “How should we care for her when she is dying?” Similarly, providers may say “dialysis is indicated” for an ICU patient in multisystem organ failure, meaning the lab values are not good and clinical guidelines indicate initiating dialysis.

Families may take this to mean that the intervention will change the patient’s outcome, or hear the word “stable” and take it as a sign the patient will survive.

“Another way to look at the situation may be to say, ‘Depending on the goals of care and whether dialysis would be of benefit, it may, or may not, be indicated,’” says Firn.

In a time-sensitive situation, says Firn, “it’s easy to just do the next indicated thing.” She says clinicians must be clear about the following things:

• the time period in which improvement can be expected after an intervention is started,
• whether a time-limited trial of a specific intervention makes sense, and
• what the plan will be if the intervention is not helping to improve or reverse the situation.

“To soften bad news, we may say the patient is not strong enough to be a candidate for a specific treatment at this time,” notes Firn. Surrogates may assume the patient will be a candidate in the future when he or she is stronger, when it is unlikely the patient will ever recover enough to receive the treatment.

Another barrier to effective communication is the sheer number of providers speaking with the family in complex cases. “Providers may essentially be communicating the same information, but using different words,” explains Firn. This could give the impression that they are not in agreement about the patient’s status. One provider may say, “The creatinine is stable today,” and another may say, “The patient still has kidney failure.”

Clinicians may default to medical terminology if they’re uncomfortable delivering bad news. “Talking more rather than less can overwhelm the surrogate with details,” says Firn. “The message we’re trying to convey could get lost.” She recommends the following practices to reduce conflicts involving prognosis:

• Make family meetings a routine part of practice.

If the team has periodic discussions with the family not only when death is imminent, but also to discuss what is good care, “that makes it less intimidating,” says Firn.

• Round informally in the ICU.

This is a good time for ethicists to ask some “bigger picture” questions of the clinical team, says Firn, such as, “What are we hoping for out of this? What is the family hoping for?”

The continual monitoring in medicine can itself be misleading. “There are so many things we are monitoring and paying attention to,” says Firm. “The numbers may be better, but have no meaningful difference in the outcome.”

The challenge is tying it all together to explain what it means for the patient’s prognosis. Bioethicists can help by asking clinicians, “What does that actually mean for the patient? Do you think it will get them off the ventilator? Do you think it will change the overall outcome?”

• Pinpoint what the family actually has to decide.

In an effort to promote autonomy, clinicians may present the situation to families as making a decision as to whether someone lives or dies. “Oftentimes it’s more about how and where,” says Firn. “We could do a better job at talking about, ‘What’s the real question here?’”

With proactive education of clinicians, fewer consultations with ethicists and the palliative care team are needed. “They will still consult ethics for serious ethical concerns,” says Firn. “But they will stop needing us for routine issues if they have a bigger toolbox to address issues upfront.”

REFERENCES

1. White DB, Ernecoff N, Buddadhumaruk P, et al. Prevalence of and factors related to discordance about prognosis between physicians and surrogate decision-makers of critically ill patients. JAMA 2016; 315(19):2086-2094.
2. White DB, Engelberg RA, Wenrich MD, et al. The language of prognostication in intensive care units. Med Decis Making 2010; 30(1):76-83.

SOURCES

• J. Randall Curtis, MD, MPH, Director, Cambia Palliative Care Center of Excellence, University of Washington, Seattle. Phone: (206) 744-3356. Fax: (206) 744-8584. Email: [email protected].
• Janice Firn, PhD, MSW, Center for Bioethics in Social Sciences and Medicine, The University of Michigan Health System, Ann Arbor. Email: [email protected].
• Douglas B. White, MD, MAS, UPMC Endowed Chair for Ethics in Critical Care Medicine/Director, Program on Ethics and Decision-Making in Critical Illness, University of Pittsburgh Medical Center. Phone: (412) 864-3757. Fax: (412) 647-8060. Email: [email protected].